Key numbers
| Measure | Value |
|---|---|
| Tests | 1790 |
| Idea variants | 261 |
| Beat benchmark | 32.2% |
| Median Sharpe vs benchmark | −0.06 |
| Verdict | No consistent edge |
All trading frequencies
Faster rules pay more costs. The table shows how that plays out across the register.
| Frequency | Tests | Beat benchmark | Median Sharpe vs bm | Verdict |
|---|---|---|---|---|
| Intraday | 347 | 18% | −0.50 | Dead end |
| Monthly | 1790 | 32% | −0.06 | No consistent edge |
| Rare (three or fewer trades per year) | 628 | 40% | −0.02 | No consistent edge |
| Weekly | 33 | 30% | −0.10 | No consistent edge |
Caveats
The 2015–2026 test window has been looked at many times during research, so we label it and do not call it a clean holdout. "Beat benchmark" compares each test with a fixed asset mix or buy-and-hold in the same window; a weak benchmark flatters the result. Many tests use US-listed ETFs.
Terms used: base rate, out-of-sample test, placebo test, dead end.
Let your AI check how ideas like yours did before it runs a single test. In our pilot, agents that checked first spent 1% of their backtests on known dead ends instead of 15%, and every backtest not wasted is time and tokens you keep. Cite this entry as ER-FRQ-MON, as of 2026-11-01.
Related evidence
Questions
Do monthly timing rules work after costs?
Monthly timing rules: 32% of 1790 tests beat their benchmark after costs in 2015–2026, median Sharpe difference −0.06. Verdict: no consistent edge.